SELECTIVE PROPERTIES OF ROUGH SPUTTERED FILMS

HAL Id: jpa-00220656

https://hal.archives-ouvertes.fr/jpa-00220656

Submitted on 1 Jan 1981

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire

HAL, est

destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

FILMS

G. Harding, S. Craig, P. Curmi, M. Lake

To cite this version:

G. Harding, S. Craig, P. Curmi, M. Lake. SELECTIVE PROPERTIES OF ROUGH SPUTTERED FILMS. Journal de Physique Colloques, 1981, 42 (C1), pp.C1-87-C1-103. �10.1051/jphyscol:1981106�.

�jpa-00220656�

JOURNAL DE PHYSIQUE

CoZZoque C l , suppZ6ment au nO1, Tome 42, janvier 1981 page Cl-87

S E L E C T I V E PROPERTIES OF ROUGH SPUTTERED FILMS

G.L. Harding, S. Craig, P. Curmi and M. Lake

P h y s i c s S c h o o l , U n i v e r s i t y of S y d n e y , N.S.W. 2006 A u s t r a l i a

Abstract.- Solar selective surfaces based on textured copper have been produced by means of two sputtering techniques. In a novel planar maqnetron cosputterina system sputter etching of bulk cop- per sheet, when seeded bv a flux of titanium atoms, led to the formation of microscogicall~ roughened surfaces with unusually complex structural formations. Preliminary sputter etchins experi- ments have also been carried out in a cylindrical magnetron sys- tem. Alternativelv, a standard planar masnetron system was used to deposit copper films of thickness 0.5 vm to 10.0 pm onto glass at sputtering pressures in the range 0.5 Pa to 100 Pa. This depo- sition technique also resulted in the formation of microscopical- ly roughened surfaces, in this case characterised by arrays of co- nes. Optical properties of both types of surface follow well de- fined trends as deposition conditions vary. High absorptance, low emittance selective surfaces have been groduced by coating homo- geneous metal carbide films onto suitably textured copper on glass films. This comcosite selective surface exhibits aood stability at temperatures up to 500°C in vacuum.

1 . Introduction.- One method of producing solar selectivity is to cre-

ate physical roughness of suitable scale on a highly reflecting metal surface. For example dendritic tunqsten /l/ and nickel /2/ deposited by CVD exhibit selectivitv fordendrite spacings % 1 pm. Strong absorp- tion occurs for wavelengths smaller than this spacing due to multiple reflections among the dendrites. The surface annears flat for longer wavelenuths, hence has relatively hiqh reflectance and low emissivity.

Sputter etchinq of bulk metal surfaces when seeded by atoms of lower sputtering yield also produces microscopically rouahened surface /3-10/.

The surface morpholouies observed can usually be classified in terms of cone, rod or ridge structures / 3 , 7 / . High solar absorptance (a) and low emittance (E) mav be obtained for surfaces with structures of suitable characteristic dimensions / 3 / . High surface temperatures produced by the ion bombardment during muttering are important in this texturing Drocess to promote surface mobility of the seed material. When the seed is sufficiently mobile, the impurity tends to collect in islands which qive rise to the various surface formations as etching proceeds.

In sectioris 2and 3 of this paper we discuss selective surfaces formed on the surface of bulk metal b~r sputter etch5ng processes carried out i n n o v e l c o s p u t t e r e t c h i n g s y s t e m s . One system allows a copper plate to

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1981106

be moved in oscillatory fashion through a planar magnetron sputter dis- charge while exposed to a flux of titanium seed, with resultant micros- copic roughening of its surface. Sputter etching is also beinq studied in a cylindrical magnetron sputterins system suitable for production of selective surfaces on long metal strips and tubes. An alternative me- thod of producing textured metal surfaces is suagested by the work of Thornton /12-14/ who obtained significant variations in surface tonogra- phy for thick copper films (25-250 pm) sputtered at low argon pressures

(0.13 and 3.9 Pa) onto substrates maintained at temneratures in the 20°C to 1 0 0 0 ~ ~ range. Thornton suggests that the structures arose from the effects of three qrowth processes : shadowinq by high points on the surface, diffusion of atoms on the surface and recrystallisation and qrain growth at the higher temperatures.

In section 4 of this paper we discuss the effects of film thickness and sputter gas pressure on copper coatinqs denosited onto glass substrates in a planar maqnetron sguttering system, with the aim of developing selective surfaces for all qlass tubular evacuated collectors. For mass

~roduction, rapid deposition of surfaces reauires that relatively thin films be deposited, and the difficulty of controlling glass tube tempe- rature makes room temperature deposition preferable. Hence film thick- nesses 0.5

-

10 pm deposited on to room temperature glass substrates have been studied for an extended range of arqon pressures 0.5

-

^{100 Pa}

in the sputtering system.

2. Planar maanetron snutter etching.- The snuttering svstem is shown schematically in figure 1. A planar maanetron with titanium electrode was used to seed the sample which consisted of copper sheet 50 mm X 80 mm X 1.0 mm. The copper sheet was sunported in a frame which could be made to translate to left on right alons a horizontal rail by a motor outside the vacuum system. The frame was coupled to the motor via a tapped collar on a threaded rod and a rotary fedd through the vacuum chamber wall. An assembly of bar magnets underneath the copper sample established a magnetron sputter zone in the form of a rectanqular track

(Fig.2.).

No cooling system was employed for the copper sample as high temperatures promote titanium seed migration durinq sputtering. Measurements showed that durins soutterinq of the copper, the temperature rose steeply in the first three to five minutes, then levelled off to an equilibrium value depending on the sputtering power (Fig.3). For the same sputte- ring power, the equilibrium temperatures reached varied by

+

50°C for the samples studied due to variations in emittance of the etched surfa- ce for the different snecirnens.

Fig.1.- Sectional view of the planar magnetron CO-sputter etching system: (A) Pirani pressure gauge ; (B) Penning pressure gauge ;

(C) High voltage feedthrouah for titanium seed electrode ; (D) grounded copper shield surrounding the titanium electrode assem- bly ; (E) 'pot' magnet for titanium planar maqnetron ; (F) Groun- ded shutter ; (G) Linear feedthrough for shutter translation ;

(H) Threaded drive shaft ; (I) Copper sample in sample holder attached to threaded shaft ; (J) Magnet assembly for copper maune- tron ; (K) Argon inlet ; (L) High voltage feedthrough for copper sample ; (F) Rotary feedthroush coupling d.c. motor to pulley

system ; ( N ) Belt and pulley svstem coupling shaft to feedthrouah ;

( 0 ) Titanium target : (P) Insulating mica support for target ;

(0) Glass vacuum chamber ; (R) Pumping port.

Fig.2.- Copper samnle showing the re- gion of plasma bombardment for statio- nary sample (hatched) and the region etched durinq continuous oscillatory CO-sputter etching (enclosed in bro- ken line).

Fig.3.- Tynical curves of copper sample temperature (T) vs. sputtering time (t) for various etching currents. Variations in temperature of f 50°C were observed for the plateaux.

A procedure was developed empirically for nroduction of acceptable se- lective surfaces on stationary copper samples /11/. The procedure in- volved sputtering the copper at 200 mA for 120s to preheat it to % 70Cf'C in the absence of titanium seed, secondly the copper was coated with a 30 nm of titanium in 6 0 S. Finally the copper electrode was sputte- red at 200 mA for 60-90s while titanium was deposited at % 0.50 nms-l.

Argon at pressure 3.0 Pa was used in all sputtering processes. It is in the final stase that the etching occurred and selective surfaces with a = 0.90 + 0.04 and room temperature emittance E = 0.11 f. 0.03 were produced in the 6mm wide sputter zone (Fiq.2).

Reflectance of a typical surface with a a 0.92, and E 0.08 is shown in figure 4. The topography of surfaces textured in these experiments resembled coiled ridges (Fig.5a). Similar structures have been observed by other workers / 7 , 9/. rieqions of discrete needles were sometimes intersnersed amonq the ridqes (Fiq.5b).

F i g . 4.- R e f l e c t a n c e ( R ) wavelenght ( A ) f o r two t e x t u r e d copper samples. S o l i d l i n e : t o t a l r e f l e c t a n c e , s h o r t dashed l i n e : s p e c u l a r r e f l e c t a n c e f o r a s t a t i o n a r y sample t r e a t e d w i t h a t i t a n i u m d e p o s i t i o n r a t e o f 0.50 nm.s-l f o r a 60s t i m e e x p o s u r e i n t h e f i n a l s t e p . a = 0 . 9 2 , E = 0.08 f o r t h i s s p e c i - men. Lonq dashed l i n e : t o t a l r e f l e c t a n c e , d o t t e d l i n e ; spe- c u l a r r e f l e c t a n c e f o r a mobil sample t r e a t e d w i t h a t i t a n l u m d e p o s i t i o n r a t e o f 0.38 nm.s-' i n t h e f i n a l , s t e p . u = 0.95,

E = 0.20 f o r t h i s specimen.

Fig.5.- E l e c t r o n micrographs showing s u r f a c e morphology f o r s t a t i o n a r y s p u t t e r e t c h e d samples. (a) C o i l e d r i d g e s . ( b ) Needles o c c a s i o n a l l y i n t e r s p e r s e d among r i d a e s .

T r a n s l a t i o n o f t h e specimen i n o s c i l l a t o r y f a s h i o n h a s been u s e d t o pro- duce uniform a r e a s o f s e l e c t i v e s u r f a c e i n a r e g i o n 50 mm ^X 45 mm ( F i g . 2 ) . A p r o c e d u r e s i m i l a r t o t h a t f o r s t a t i o n a r y specimens was used, how-

ever in the final (sputter etching) stage, a sputtering time of 300s and various titanium deposition rates (0.14 nms-l to 0.50 nms-l) were used. Good selective properties were obtained, higher a and E values being associated with the higher titanium seed deposition rates.

Figure 4 shows the reflectance for a specimen with a = 0.95 and E = 0.20, produced using a titanium deposition rate of 0.38 nms-l. Electron micro- g r a ~ h s of this surface (Fig.6) show an unusually complex ridge struc- ture.

Fig.6.- Electron micrographs showing surface morpnology for a sample translated throuqh the soutter discharge with tita- nium deposition rate 0.38 nm.s-l.

A simplified etching procedure involved translating the sample in os- cillatory fashion, sputtering it continuously& 200 mA and continuous- ly depositinq seed at 0.38 nms-'

,

0.44 nms-' or 0.50 nms-'

.

^Etching

times were varied between 5 and 12 minutes.

The selective surfaces produced were not perfectly reproducible, how- ever some trends in a and E are clear. Figure 7 shows a and E for tita- nium deposition rates 0.38, 0.44 and 0.50 nms-l, as a function of to- tal sputtering time. The hiqhest titaniumdeposition rates resulted in hiqhest emittance (exce~t for a few anomalous specimens). Optimum selec- tivity was obtained for the 0.44 runs-' rate. Figure 8 shows reflectance vs wavelength for two specimens. A wide ranse of surface morphologies was obtained for this production technique /11/.

Samples of sputter etched surface were enclosed in a continuously pum- ped vacuum chamber and subjected to a temperature of 500°C for > 4,000 hours. Figure 9 shows a and E VS annealing time for a specimen of ini- tial absorptance 0.95 and initial emittance 0.20. Both a and E decrease for the first 2,000 hours, then appear to become stable. The final ab-

sorptance is a ^% 0.88 and final emittance 0.09. It is not clear whe- ther the structure of the surface changes during annealing at this tem- perature. A possible mechanism for the changes is slow oxidation of residual titanium on the surface due to extremelv low partial pressures of oxygen or H 2 0 in the vacuum chamber.

Fiq.7.- Absorptance (a) and emittance ( E ) =.CO-sputtering time (t) for various titanium seed deposition rates. Samples were etched by the continuous CO-sputter etching procedure.

Solid circles : 0.38 nm.s-l ; solid squares : 0.44 nm.s-l ;

open circles : 0.50 nm.s-l.

Fig. R.

-

Reflectance (R) =.wavelength (X) for samples produced by the continuous CO-sputter etchinq procedure. Solid line :

total reflectance, short dashed line : specular reflectance for titanium deposition rate 0.44 nm.sql, total sputtering time 7.5. min. a = 0.90, E = 0.11 for this specimen. Long dashed line:

total reflectance, dotted line : specular reflectance for tita- nium deposition rate 0.50 nrn.se1, total sputtering time 7 min.

a = 0.93, E = 0.37 for this specimen.

Fig. 9.

-

Absorptance (a) and emittance ( E ) vs. ageina time

OO 1000 2KlO 3000 LOOO for a sputter etched copper

surface annealed in vacuum at ANNEALING TIME (h) 500°C.

3. Sputter etching in a cylindrical magnetron.- Sputter etchinq experi- ments are now being carried out in a cylindrical magnetron sputtering system suitable for production of selective surfaces on long metal pla- tes and metal tubes. Figure 10 shows a schematic diagram of the coating 1

system.

Fig.10.- Sectional view of a cylindri- cal magnetron sputter etchinq system.

A : Vacuum chamber ; B : Titanium elec-

H

trode ; C : Metal plate for sputter et-

/

^{chinq ; D} : Rotary feedthrough (and high voltaqe feedthrough) ; E : Metal tubes for sputter etching ; F : Screens to minimize cross contamination of sput- ter etched specimens ; G : Arqon inlet;

H : magnet coil.

A central titanium electrode provides seed material for metal plates or tubes. The titanium electrode is water cooled, however the sputtered plates or tubes lose energy only by radiation and thermal conduction through the argon gas and so reach high temperatures. Six or more plates or tubes could in principle be sputter etched simultaneously using the chamber design shown. Rotary feedthroughs allow the specimens to be ro- tated during etching. The sputter etching of copper strips, 260 mm ^X 50 mm X 0.6 mm is presently being investigated.

Selective surfaces with a Q, 0.90 and E % 0.10 have been produced with excellent uniformity along the strip. Sputtering time, seed deposition rate and sputter gas pressure are being varied in an attempt to optimise the selective surface manufacture. The etching of alternative materials such as stainless steel and nickel will also be investigated.

4. Sputtered copper films on glass substrates.- A planar magnetron was used as a high rate s~uttering source to deposit copper coatings onto glass substrates at room temperature (Fig.11).

pig.11.- Sectional view of pla- nar maqnetron sputter coating system. A : Permanent magnet H array ; B : Cathode assembly ;

C : Teflon insulator ; D : Ar- gon inlet ; E : Anode with cen- tral aperture ; F : Rotating I substrate mount ; G : Guide for

drive shafts of substrate mounts and shutter ; H,I : Pressure gauges ; J : magnetic all!^ cou- pled feedthrough for substrate mounts ; K : Pumping port ; L :

Shutter ; M : Grounded vacuum chamber.

Six glass substrates were thermally clamped to two grounded rotatinq copper platforms, enabling the substrates to be held at room temperatu- re while individually exposed to the sputtered flux through an aperture

in the anode. The chamber was initially evacuated to a base pressure of c 0.1 mPa using a four inch liquid nitrogen trapped oil diffusion pump.Deposition parameters for the sputtered copper films are listed in table I. Deposition rate, argon gas flow rate, glass substrate tem- perature and electrode-substrate distance were effectively constant for all films. Argon pressure was regulated by throttling the diffusion pump, and measured using a Pirani aauge calibrated within 2%. Different film thicknesses were obtained by varying the exposure time of the glass substrates to the sputtered flux.

Table I.- Deposition conditions for copper films deposited onto glass substrates.

Cathode OFHC Copper

Magnetic field 0.080 T

Target-substrate distance Substrate temperature

Deposition rate 0.5 m/min.

Film thickness (+10%) (pm) 0.5 1.0 3.0 10 neposition time (min) 1.0 2.0 6.0 20.0

Argon gas flow rate 2.7 x 1 0 - ~ m3s-I at STP Argon pressure (Pa) 0.65 6.5 19.5 39 65 97.5 Discharge Current (A) 0.45 0.65 1.0 1.6 2.4 3.4 Current density 280 410 630 1,000 1,500 2,200 Discharge voltage (V) 405 355 340 340 345 355

The basic surface topographical features observed were arrays of cones.

Figure 12 shows scanning electron micrographs viewed at 25' to the pla- ne of the surfaces. The micrographs illustrate the trend in surface features as sputtering pressure and film thickness is varied. The sur- face density of cones increases with sputtering pressure, ranging from the relatively featureless surfaces of films sputtered at low argon pressures to the densely packed cones of high pressure deposited films.

Cone size is proportional to film thickness, having dimensions of order 10% of the thickness. For film thickness 3 pm and 10 um, as argon pres- sures increase from 20 Pa to 100 Pa, cone size decreases corresponding to a transition from isolated clusters of cones to a densely packed array.

-eaxauy y q y M 6uypuodsaxxoa 'axnssaxd

yq y~

paseaxauy p u e quear3yubys s e

M

aa ue qa a~

~a x ploq a y q

3 0 q u a u o d u r o a asn33rp ay=) 'mr

i 0.0~

p u e m rt

0-

ssau E

-YDTYl 30 s ur T T3 x o d 'Pd 6E qe paqrsodap u rI

r 3 y 3 y y q urri e 1 T 0 3 yqbua~anr?~

aaueqaa~gax s~ SA

oy

ET s

ax n6 y~

'umxqaads xeTos a

y q u r sy q6 ua

~a ae

~ 103

axnssaxd bu~s-eaxauy Y~

TM paseaxaap sa~ueqaa~gax

pqoq p u e a~qyby16au

s e M aaueqaaTjax ~

eq oq 3 0 q ua u oi i ur o o asn33yp a

y 7 'urrt 0 . 1 o q ssauyayyq 5 .0

30 sur~y3 xoa '(ET-6rd)

xaddoa y T n q paysy~od 30

quaaxad ~ a 3 e uyyqrfi

pax-ex3uy p u e aIqysrn a y?

u y aaueqaaT3ax y

qy

~ 'xe~naads axaM ed

59.0 qe

paxaqqnds s m~

yj 'ssauyayyq 3

0 quapuadapu~ -d601oydxom

aaegxns a y q uo aa

-napuadap pa y xe u r e pade~dsyp sur~r3 xaddoa

a y q 3 0 sayqxadoxd

~earqdo a y

&

'urri 5.1 aprs buraey saxenbs a

y q 'ydex60~ayur yaea

1 0 3 ames a y q S ? uorqear3yube~

'(

saxnssaxd uobxe 6) snoyxen qe

paqysod

-ap (p) sassauyayyq snoyxen

3 0 sur~y3 x o

3 Kbo~oqdxour aae3xns

buy~oqs aa

-e31ns UTT3 a y 7 0 q . , SZ 78 paMayn

sydex6oxoym uoxqaaIa Buyuueas

-' zI -~

T~

sing density of cones of characteristic dimensions 0.3 to 1.0 um.

Fig. 13.

-

Reflectance (R) v& wavelength (X) for three snuttered copper films. Deposition conditions are listed in table I.

Solid line : Reflectance of films deposited at 0.65 Pa ; long dashes and spaced dots : 1.0 ym thick film deposited at 39 Pa ;

Short dashes and close spaced dots : 10 ym thick film deposited at 39 Pa. (Specular reflectance is indicated by dots only, dif- fuse reflectance by dashes only and total reflectance by mixed dots and dashes).

Reflectances were completely diffuse over most of the solar spectrum for films of thickness 3.0 pm and 10.0 ym manufactured at 39, 65 and 97 Pa. Reflectances of a 10 um thick film deposited at 39 Pa are shown in figure 13. A general trend of decreasinq total reflectance for wave- lengths in the solar spectrum as sputtering gas pressure increases is observed for all four film thicknesses and this is shown in figure 14 for 3 ym thick films.

Fig. 14.- Total reflectance (R) =.wavelength ( A ) for 3 um thick sputtered copper films. Deposition conditions are listed in table I. Solid line : film deposited at 0.65 Pa ;

Solid line with dots : film deposited at 6.5 Pa ; long dashed line : film deposited at 19.5 Pa ; long dashed line with dots ;

film deposited at 39 Pa ; Short dashed line : film deposited

at 65 Pa ; Short dashed line with dots : film deposited at 97.5 Pa.

Figures 15 to 18 illustrate the trends in absorptance a and room tem- perature emittance E for varying film thickness and sputtering pressu- re. Absorptance and emittance increase with film thickness and sputte- ring pressure. Most absorptances and emittances were reproducible with- in a few percent.

Fig. 16.- Absorptance (a) and emittance ( E ) at 300K

E.

sputtering gas pressure for copper film thickness 1.0 um.

1.0

.8

.6

a,€

-

d=O.Spm

-

-

- I 1

. _ _ _ _ _ _ . _ - - - ^-

0 ^{, , I l l}

20 40 60 80 100

PRESSURE (Pa)

Fig. 15.- Absorptance ( a ) and emittance ( E ) at 300K vs.

sputtering gas pressure for copper film thickness 0 . 5 pm.

,

20 40 60 80 100

PRESSURE (Pal

Fig. 17.

-

Absorptance ( a ) and emittance ( E ) at 300K 5 sputtering gas pressure for copper film thickness 3.0 um.

/ .

01- I ' ^{" I I}

20 40 60 80 100

PRESSURE (Pa)

Fig. 18.- Absorptance (a) and emittance ( E ) at 300 K vs.

sputtering gas pressure for copper film thickness 10.0 vm.

Optimum values of a/& are obtained for the 1.0 thick films deposited at 39 Pa. High solar absorptances and moderately low emittances have

been obtained by coating a reactively sputtered homogeneous metal carbi- de film on these films.

Absorptances a ^Q 0.90 and E % 0.04 to 0.05 were obtained compared to a Q 0.80 and E Q 0.02 for a similar interference layer on smooth copper deposited at 0.65 Pa. Figure 19 shows reflectance =wavelength for a metal carbide deposited onto smooth and textured films. The latter films have been aged at 200°C, 300°C, 400°C and 500°C in a continuously pum- ped vacuum chamber at pressure < 1 mPa. Ageing results are summarised in table 11.

Table 11.- Solar absorptance a and emittance E (at 300K) before and after ageing for homogeneOus stainless steel carbide films on textu- red copper.

Ageing Temperature Total time (h) 0

a 0.895

E 0.053

Ageing Temperature Total time (h) 0

a 0.90

E 0.054

Ageing Temperature Total time (h) 0

CL 0.90

E 0.052

Ageing Temperature Total time (h) 0

a 0.90

E 0.035

The absorptances are stable at 300°C and 400'~ while emittances decrea- se initially. At 500°C both absorptance and emittance increase slowly possibly due to diffusion at the metal carbidecopprinterface / 1 6 / . The anomalous decrease in absorptance during the 200°C anneal may be asso- ciated with reaction of the metal carbide surface with residual gas in the continuously pumped chamber.

Fig. 19.

-

^Ref lectance. (R) wavelength ( h ) for homogeneous metal carbide layers overlayed on copper films sputtered at 0.65 Pa (smooth surfaces) and 39 Pa (textured surfaces)

.

Solid line : Copper deposited at 0.65 Pa ; solid line with dots ; Copper deposited at 39 Pa ; Dashed line : Metal car- bide overlayed on 0.65 Pa copper ; Dashed line with dots :

Metal carbide overlayed on 39 Pa copper.

5. Conclusion.- Sputter etching of copper using a planar magnetron sys- tem yielded good selective surfaces with moderate stability at 500°C.

However this etching apparatus does not seem easily adaptable to produ- cing large areas of selective surface.

The cylindrical magnetron technology offers greater prospects in this regard, however work at Sydney University is still in a preliminary stage. Sputter etched strip or pipe may form a suitable receiver ele- ment in vacuum insulated glass envelopes incorporated in parabolic

trough concentrators or CPCs designed for production of high tempera- tures (300

-

500°C). Alternative materials to copper such as stainless steel and nickel may offer greater stability at high temperatures in vacuum or in air. Coating of these surfaces with a refractory material may also improve the stability of the surface.

Selective surfaces based on textured thin copper films deposited on glass in high argon pressures offer an interesting prospect for all- glass evacuated collectors. However, as themechanism of formation of these surfaces is unclear, it is not known whether mass production of the textured films is feasible.

Acknowledgements.- The authors would like to thank their colleagues Dr.

C. M. Horwitz, Dr. R. McPhedran and Dr. D. McKenzie for advice and en- couragement. This work was funded by grants from the Sydney University Energy Research Centre, the Government of New South Wales and Saudi Arabian interests under the auspices of the University of Sydney ~ c i 8 n - ce Foundation.

References

Cuomo,J.J., Ziegler,J.F. and W0odwal1,J.M.~ Appl.

Phys. Lett.

26

( 1 9 7 5 ) 5 5 7

Grimmer,D.P., Hern,K.C. and McCreary,W.J., J. Vac.

Sci. Technol.

15

( 1 9 7 8 ) 59

Berg,R.S. and Kominiak,G.J., J. Vac. Sci. Technol.

1 3 ( 1 9 7 5 ) 403

-

Witcombe,M.J., J. Appl. Phys.

46

( 1 9 7 5 ) 5 0 5 3 Vossen, J.L., J. Phys. E.

12

( 1 9 7 9 ) 1 5 9

Wehner,G.K. and Hajicek,D.J., J. Appl. Phys.

42

( 1 9 7 1 ) 1 1 4 5

Hudson,W.R., J. Vac. Sci. Technol.

30

( 1 9 7 7 ) 286 Wehner, G., J. Appl. Phys.

30

( 1 9 5 9 ) 1 7 6 2

Oohashi,T. and Yamanka,S., Japan J. Appl. Phys.

11

( 1 9 7 2 ) 1 5 8 1

Stewart,A.D.G. and Thompson,M.W., J. Mater. Sci.

4 ( 1 9 6 9 ) 5 6

-

Curmi,P. and Harding,G.L., .To be published in J. Vac. Sci.

Technol. (Nov. 1 9 0 0 ) .

Thornton, J.A., J. Vac. Sci. Technol.

11

( 1 9 7 4 ) 666 Thornton,J.A., J. Vac. Sci. Technol.

12

( 1 9 7 5 ) 830 Thornton,J..A., Proc. A.E.S. Coatings for Solar Collectors Symposium. Atlanta, Georgia ( 1 9 7 6 ) 63 IIarding,G.L. and Craig,S., J. Vac. Sci. Technol.

16

( 1 9 7 9 ) 857

/16/ Harding, G.L., Sol. Eneray Materials,

2 ( 1 9 8 0 ) 469

-